第一次测试提交

Megatron-LM-core_r0.7.0.beta/megatron/core/models/bert/bert_layer_specs.py

+from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
+from megatron.core.fusions.fused_layer_norm import FusedLayerNorm
+from megatron.core.tensor_parallel.layers import ColumnParallelLinear, RowParallelLinear
+from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
+from megatron.core.transformer.custom_layers.transformer_engine import (
+    TEDotProductAttention,
+    TELayerNormColumnParallelLinear,
+    TERowParallelLinear,
+)
+from megatron.core.transformer.dot_product_attention import DotProductAttention
+from megatron.core.transformer.enums import AttnMaskType
+from megatron.core.transformer.identity_op import IdentityOp
+from megatron.core.transformer.mlp import MLP, MLPSubmodules
+from megatron.core.transformer.spec_utils import ModuleSpec
+from megatron.core.transformer.transformer_layer import TransformerLayer, TransformerLayerSubmodules
+
+# Use this spec to use lower level Transformer Engine modules (required for fp8 training)
+bert_layer_with_transformer_engine_spec = ModuleSpec(
+    module=TransformerLayer,
+    submodules=TransformerLayerSubmodules(
+        self_attention=ModuleSpec(
+            module=SelfAttention,
+            params={"attn_mask_type": AttnMaskType.padding},
+            submodules=SelfAttentionSubmodules(
+                linear_qkv=TELayerNormColumnParallelLinear,
+                core_attention=TEDotProductAttention,
+                linear_proj=TERowParallelLinear,
+                q_layernorm=IdentityOp,
+                k_layernorm=IdentityOp,
+            ),
+        ),
+        self_attn_bda=get_bias_dropout_add,
+        mlp=ModuleSpec(
+            module=MLP,
+            submodules=MLPSubmodules(
+                linear_fc1=TELayerNormColumnParallelLinear, linear_fc2=TERowParallelLinear,
+            ),
+        ),
+        mlp_bda=get_bias_dropout_add,
+    ),
+)
+
+# Use this spec for an implementation using only modules in megatron core
+bert_layer_local_spec = ModuleSpec(
+    module=TransformerLayer,
+    submodules=TransformerLayerSubmodules(
+        input_layernorm=FusedLayerNorm,
+        self_attention=ModuleSpec(
+            module=SelfAttention,
+            params={"attn_mask_type": AttnMaskType.padding},
+            submodules=SelfAttentionSubmodules(
+                linear_qkv=ColumnParallelLinear,
+                core_attention=DotProductAttention,
+                linear_proj=RowParallelLinear,
+                q_layernorm=IdentityOp,
+                k_layernorm=IdentityOp,
+            ),
+        ),
+        self_attn_bda=get_bias_dropout_add,
+        pre_mlp_layernorm=FusedLayerNorm,
+        mlp=ModuleSpec(
+            module=MLP,
+            submodules=MLPSubmodules(
+                linear_fc1=ColumnParallelLinear, linear_fc2=RowParallelLinear,
+            ),
+        ),
+        mlp_bda=get_bias_dropout_add,
+        sharded_state_dict_keys_map={
+            'input_layernorm.': 'self_attention.linear_qkv.layer_norm_',
+            'pre_mlp_layernorm.': 'mlp.linear_fc1.layer_norm_',
+        },
+    ),
+)

Megatron-LM-core_r0.7.0.beta/megatron/core/models/bert/bert_lm_head.py

+import torch
+from torch import Tensor
+
+from megatron.core.fusions.fused_layer_norm import FusedLayerNorm
+from megatron.core.transformer.module import MegatronModule
+from megatron.core.transformer.transformer_config import TransformerConfig
+from megatron.core.transformer.utils import get_linear_layer
+
+
+class BertLMHead(MegatronModule):
+    """Masked LM head for Bert.
+
+    Args:
+        hidden_size: hidden size
+        config (TransformerConfig): TransformerConfig object
+     """
+
+    def __init__(
+        self, hidden_size: int, config: TransformerConfig,
+    ):
+        super().__init__(config=config)
+
+        # TODO: Should switch this to TE ?
+        self.dense = get_linear_layer(
+            hidden_size, hidden_size, config.init_method, config.perform_initialization
+        )
+
+        setattr(self.dense.weight, 'sequence_parallel', config.sequence_parallel)
+        setattr(self.dense.bias, 'sequence_parallel', config.sequence_parallel)
+
+        self.layer_norm = FusedLayerNorm(
+            config=config, hidden_size=hidden_size, eps=config.layernorm_epsilon,
+        )
+
+        self.gelu = torch.nn.functional.gelu
+
+    def forward(self, hidden_states: Tensor) -> Tensor:
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.gelu(hidden_states)
+        hidden_states = self.layer_norm(hidden_states)
+        return hidden_states

Megatron-LM-core_r0.7.0.beta/megatron/core/models/bert/bert_model.py

+# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
+import os
+from collections import OrderedDict
+from typing import Dict, Literal, Optional
+
+import torch
+from torch import Tensor
+
+from megatron.core import parallel_state, tensor_parallel
+from megatron.core.dist_checkpointing.mapping import ShardedStateDict
+from megatron.core.models.bert.bert_lm_head import BertLMHead
+from megatron.core.models.bert.pooler import Pooler
+from megatron.core.models.common.embeddings.language_model_embedding import LanguageModelEmbedding
+from megatron.core.models.common.embeddings.rotary_pos_embedding import RotaryEmbedding
+from megatron.core.models.common.language_module.language_module import LanguageModule
+from megatron.core.transformer.enums import AttnMaskType, ModelType
+from megatron.core.transformer.spec_utils import ModuleSpec
+from megatron.core.transformer.transformer_block import TransformerBlock
+from megatron.core.transformer.transformer_config import TransformerConfig
+from megatron.core.transformer.utils import get_linear_layer
+from megatron.core.utils import make_tp_sharded_tensor_for_checkpoint
+
+
+class BertModel(LanguageModule):
+    """Transformer language model.
+
+    Args:
+        config (TransformerConfig): transformer config
+        num_tokentypes (int) : Set to 2 when args.bert_binary_head is True, and 0 otherwise. Defaults to 0.
+        transformer_layer_spec (ModuleSpec): Specifies module to use for transformer layers
+        vocab_size (int): vocabulary size
+        max_sequence_length (int): maximum size of sequence. This is used for positional embedding
+        pre_process (bool): Include embedding layer (used with pipeline parallelism)
+        post_process (bool): Include an output layer (used with pipeline parallelism)
+        parallel_output (bool): Do not gather the outputs, keep them split across tensor parallel ranks
+        share_embeddings_and_output_weights (bool): When True, input embeddings and output logit weights are shared. Defaults to False.
+        position_embedding_type (string): Position embedding type. Options ['learned_absolute', 'rope'].
+            Defaults is 'learned_absolute'.
+        rotary_percent (float): Percent of rotary dimension to use for rotary position embeddings.
+            Defaults to 1.0 (100%). Ignored unless position_embedding_type is 'rope'.
+    """
+
+    def __init__(
+        self,
+        config: TransformerConfig,
+        num_tokentypes: int,
+        transformer_layer_spec: ModuleSpec,
+        vocab_size: int,
+        max_sequence_length: int,
+        pre_process: bool = True,
+        post_process: bool = True,
+        fp16_lm_cross_entropy: bool = False,
+        parallel_output: bool = True,
+        share_embeddings_and_output_weights: bool = False,
+        position_embedding_type: Literal['learned_absolute', 'rope'] = 'learned_absolute',
+        rotary_percent: float = 1.0,
+        seq_len_interpolation_factor: Optional[float] = None,
+        add_binary_head=True,
+        return_embeddings=False,
+    ):
+        super(BertModel, self).__init__(config=config)
+
+        if return_embeddings:
+            assert self.post_process and self.add_binary_head
+
+        assert (
+            os.getenv('NVTE_ALLOW_NONDETERMINISTIC_ALGO') == '0'
+            or os.getenv('NVTE_FLASH_ATTN') == '0'
+        ), "Bert currently does not support flash attention. Please set env variable NVTE_FLASH_ATTN=0 or set NVTE_ALLOW_NONDETERMINISTIC_ALGO=0"
+
+        self.config: TransformerConfig = config
+        self.transformer_layer_spec: ModuleSpec = transformer_layer_spec
+        self.vocab_size = vocab_size
+        self.max_sequence_length = max_sequence_length
+        self.pre_process = pre_process
+        self.post_process = post_process
+        self.fp16_lm_cross_entropy = fp16_lm_cross_entropy
+        self.parallel_output = parallel_output
+        self.share_embeddings_and_output_weights = share_embeddings_and_output_weights
+        self.position_embedding_type = position_embedding_type
+        self.add_binary_head = add_binary_head
+        self.return_embeddings = return_embeddings
+
+        # megatron core pipelining currently depends on model type
+        self.model_type = ModelType.encoder_or_decoder
+
+        # Embeddings.
+        if self.pre_process:
+            self.embedding = LanguageModelEmbedding(
+                config=self.config,
+                vocab_size=self.vocab_size,
+                max_sequence_length=self.max_sequence_length,
+                position_embedding_type=position_embedding_type,
+                num_tokentypes=num_tokentypes,
+            )
+
+        if self.position_embedding_type == 'rope':
+            self.rotary_pos_emb = RotaryEmbedding(
+                kv_channels=self.config.kv_channels,
+                rotary_percent=rotary_percent,
+                rotary_interleaved=self.config.rotary_interleaved,
+                seq_len_interpolation_factor=seq_len_interpolation_factor,
+            )
+
+        # Transformer.
+        self.encoder = TransformerBlock(
+            config=self.config,
+            spec=self.transformer_layer_spec,
+            pre_process=self.pre_process,
+            post_process=self.post_process,
+        )
+
+        # Output
+        if post_process:
+            # TODO: Make sure you are passing in the mpu_vocab_size properly
+            self.lm_head = BertLMHead(config.hidden_size, config,)
+
+            self.output_layer = tensor_parallel.ColumnParallelLinear(
+                config.hidden_size,
+                self.vocab_size,
+                config=config,
+                init_method=config.init_method,
+                bias=True,
+                skip_bias_add=False,
+                gather_output=not self.parallel_output,
+                skip_weight_param_allocation=pre_process and share_embeddings_and_output_weights,
+            )
+
+            self.binary_head = None
+            if self.add_binary_head:
+                # TODO: Shoudl switch this to TE ?
+                self.binary_head = get_linear_layer(
+                    config.hidden_size, 2, config.init_method, config.perform_initialization
+                )
+
+                self.pooler = Pooler(
+                    config.hidden_size, config.init_method, config, config.sequence_parallel
+                )
+
+        if self.pre_process or self.post_process:
+            self.setup_embeddings_and_output_layer()
+
+    def bert_extended_attention_mask(self, attention_mask: Tensor) -> Tensor:
+        """Creates the extended attention mask
+
+        Converts the attention mask of dimension [batch size, 1, seq len] to [batch size, 1, seq len, seq len] and makes it binary
+
+        Args:
+            attention_mask (Tensor): The input attention mask
+
+        Returns:
+            Tensor: The extended binary attention mask
+        """
+        # We create a 3D attention mask from a 2D tensor mask.
+        # [b, 1, s]
+        attention_mask_b1s = attention_mask.unsqueeze(1)
+        # [b, s, 1]
+        attention_mask_bs1 = attention_mask.unsqueeze(2)
+        # [b, s, s]
+        attention_mask_bss = attention_mask_b1s * attention_mask_bs1
+        # [b, 1, s, s]
+        extended_attention_mask = attention_mask_bss.unsqueeze(1)
+
+        # Convert attention mask to binary:
+        extended_attention_mask = extended_attention_mask < 0.5
+
+        return extended_attention_mask
+
+    def bert_position_ids(self, token_ids):
+        # Create position ids
+        seq_length = token_ids.size(1)
+        position_ids = torch.arange(seq_length, dtype=torch.long, device=token_ids.device)
+        position_ids = position_ids.unsqueeze(0).expand_as(token_ids)
+
+        return position_ids
+
+    def set_input_tensor(self, input_tensor: Tensor) -> None:
+        """Sets input tensor to the model.
+
+        See megatron.model.transformer.set_input_tensor()
+
+        Args:
+            input_tensor (Tensor): Sets the input tensor for the model.
+        """
+        # This is usually handled in schedules.py but some inference code still
+        # gives us non-lists or None
+        if not isinstance(input_tensor, list):
+            input_tensor = [input_tensor]
+
+        assert len(input_tensor) == 1, 'input_tensor should only be length 1 for gpt/bert'
+        self.encoder.set_input_tensor(input_tensor[0])
+
+    def forward(
+        self,
+        input_ids: Tensor,
+        attention_mask: Tensor,
+        tokentype_ids: Tensor = None,
+        lm_labels: Tensor = None,
+        inference_params=None,
+    ):
+        """Forward function of BERT model
+
+        Forward function of the BERT Model This function passes the input tensors
+        through the embedding layer, and then the encoder and finally into the post
+        processing layer (optional).
+
+        It either returns the Loss values if labels are given  or the final hidden units
+        """
+        extended_attention_mask = self.bert_extended_attention_mask(attention_mask)
+
+        if parallel_state.is_pipeline_first_stage():
+            input_ids = input_ids
+            position_ids = self.bert_position_ids(input_ids)
+        else:
+            position_ids = None
+            input_ids = None
+
+        # Encoder embedding.
+        if self.pre_process:
+            encoder_input = self.embedding(
+                input_ids=input_ids, position_ids=position_ids, tokentype_ids=tokentype_ids
+            )
+        else:
+            # intermediate stage of pipeline
+            # encoder will get hidden_states from encoder.input_tensor
+            encoder_input = None
+
+        # Rotary positional embeddings (Why not move this into BERT/GPTEmberdding ?)
+        rotary_pos_emb = None
+        if self.position_embedding_type == 'rope':
+            rotary_seq_len = self.rotary_pos_emb.get_rotary_seq_len(
+                inference_params, self.encoder, encoder_input, self.config
+            )
+            rotary_pos_emb = self.rotary_pos_emb(rotary_seq_len)
+
+        # Run encoder.
+        hidden_states = self.encoder(
+            hidden_states=encoder_input,
+            attention_mask=extended_attention_mask,
+            inference_params=inference_params,
+            rotary_pos_emb=rotary_pos_emb,
+        )
+        if not self.post_process:
+            return hidden_states
+
+        if self.add_binary_head:
+            pooled_output = self.pooler(hidden_states, 0)
+
+        if self.return_embeddings:
+            embeddings = torch.transpose(hidden_states, 0, 1)
+            masks = torch.sum(attention_mask, dim=1)
+            # Collect masked embeddings.
+            output = torch.zeros(
+                size=(embeddings.shape[0], embeddings.shape[2]),
+                dtype=torch.float32,
+                device=torch.cuda.current_device(),
+            )
+            for i, (embedding, mask) in enumerate(zip(embeddings, masks)):
+                output[i, :] = torch.mean(embedding[1 : mask - 1], dim=0)
+            return output
+
+        # logits and loss
+        output_weight = None
+        if self.share_embeddings_and_output_weights:
+            output_weight = self.shared_embedding_or_output_weight()
+
+        hidden_states_after_lm_head = self.lm_head(hidden_states=hidden_states)
+        logits, _ = self.output_layer(hidden_states_after_lm_head, weight=output_weight)
+
+        binary_logits = None
+        if self.binary_head is not None:
+            binary_logits = self.binary_head(pooled_output)
+
+        if lm_labels is None:
+            # [s b h] => [b s h]
+            return logits.transpose(0, 1).contiguous(), binary_logits
+
+        loss = self.compute_language_model_loss(lm_labels, logits)
+
+        return loss, binary_logits

Megatron-LM-core_r0.7.0.beta/megatron/core/models/bert/pooler.py

+import torch
+from torch import Tensor
+
+from megatron.core import tensor_parallel
+from megatron.core.transformer.module import MegatronModule
+from megatron.core.transformer.transformer_config import TransformerConfig
+from megatron.core.transformer.utils import get_linear_layer
+
+
+class Pooler(MegatronModule):
+    """Pooler layer.
+
+    Pool hidden states of a specific token (for example start of the
+    sequence) and add a linear transformation followed by a tanh.
+
+    Args:
+        hidden_size (int): The hidden size_
+        init_method (callable): weight initialization method for the linear layer. bias is set to zero.
+        config (TransformerConfig): The transformer configuration
+        sequence_parallel (bool): Using squence parallel ? Defaults to False
+    """
+
+    def __init__(
+        self,
+        hidden_size: int,
+        init_method: callable,
+        config: TransformerConfig,
+        sequence_parallel: bool = False,
+    ):
+        super(Pooler, self).__init__(config)
+        # TODO: Shoudl switch this to TE ?
+        self.dense = get_linear_layer(
+            hidden_size, hidden_size, init_method, config.perform_initialization
+        )
+        self.sequence_parallel = sequence_parallel
+
+    def forward(self, hidden_states: Tensor, sequence_index=0):
+        # hidden_states: [s, b, h]
+        # sequence_index: index of the token to pool.
+
+        # gather data along sequence dimensions
+        # same pooler is run on all tensor parallel nodes
+        if self.sequence_parallel:
+            hidden_states = tensor_parallel.gather_from_sequence_parallel_region(
+                hidden_states, tensor_parallel_output_grad=False
+            )
+
+        pooled = hidden_states[sequence_index, :, :]
+        pooled = self.dense(pooled)
+        pooled = torch.tanh(pooled)
+        return pooled

Megatron-LM-core_r0.7.0.beta/megatron/core/models/common/embeddings/language_model_embedding.py

+# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
+
+from typing import Literal
+
+import torch
+from torch import Tensor
+
+from megatron.core import tensor_parallel
+from megatron.core.transformer.module import MegatronModule
+from megatron.core.transformer.transformer_config import TransformerConfig
+
+
+class LanguageModelEmbedding(MegatronModule):
+    """Language model embeddings.
+
+    Args:
+        config (TransformerConfig): config object with all necessary configs for TransformerBlock
+        vocab_size (int): vocabulary size
+        max_sequence_length (int): maximum size of sequence. This
+                             is used for positional embedding
+        add_position_embedding (bool): Add a position embedding.
+        embedding_dropout_prob (float): dropout probability for embeddings
+        num_tokentypes (int): Set to 0 without binary head, and 2 with a binary head . Defaults to 0.
+    """
+
+    def __init__(
+        self,
+        config: TransformerConfig,
+        vocab_size: int,
+        max_sequence_length: int,
+        position_embedding_type: Literal['learned_absolute', 'rope'] = 'learned_absolute',
+        num_tokentypes: int = 0,
+    ):
+        super().__init__(config=config)
+
+        self.config: TransformerConfig = config
+        self.vocab_size: int = vocab_size
+        self.max_sequence_length: int = max_sequence_length
+        self.add_position_embedding: bool = position_embedding_type == 'learned_absolute'
+        self.num_tokentypes = num_tokentypes
+
+        # Word embeddings (parallel).
+        self.word_embeddings = tensor_parallel.VocabParallelEmbedding(
+            num_embeddings=self.vocab_size,
+            embedding_dim=self.config.hidden_size,
+            init_method=self.config.init_method,
+            config=self.config,
+        )
+
+        # Position embedding (serial).
+        if self.add_position_embedding:
+            self.position_embeddings = torch.nn.Embedding(
+                self.max_sequence_length, self.config.hidden_size
+            )
+
+            # Initialize the position embeddings.
+            if self.config.perform_initialization:
+                self.config.init_method(self.position_embeddings.weight)
+
+        if self.num_tokentypes > 0:
+            self.tokentype_embeddings = torch.nn.Embedding(
+                self.num_tokentypes, self.config.hidden_size
+            )
+            # Initialize the token-type embeddings.
+            if self.config.perform_initialization:
+                self.config.init_method(self.tokentype_embeddings.weight)
+        else:
+            self.tokentype_embeddings = None
+
+        # Embeddings dropout
+        self.embedding_dropout = torch.nn.Dropout(self.config.hidden_dropout)
+
+    def zero_parameters(self):
+        """Zero out all parameters in embedding."""
+        self.word_embeddings.weight.data.fill_(0)
+        self.word_embeddings.weight.shared = True
+        self.position_embeddings.weight.data.fill_(0)
+        self.position_embeddings.weight.shared = True
+        if self.num_tokentypes > 0:
+            self.tokentype_embeddings.weight.data.fill_(0)
+            self.tokentype_embeddings.weight.shared = True
+
+    def forward(self, input_ids: Tensor, position_ids: Tensor, tokentype_ids: int = None) -> Tensor:
+        """Forward pass of the embedding module.
+
+        Args:
+            input_ids (Tensor): The input tokens
+            position_ids (Tensor): The position id's used to calculate position embeddings
+            tokentype_ids (int): The token type ids. Used when args.bert_binary_head is set to True. Defaults to None
+
+        Returns:
+            Tensor: The output embeddings
+        """
+        word_embeddings = self.word_embeddings(input_ids)
+        if self.add_position_embedding:
+            position_embeddings = self.position_embeddings(position_ids)
+            embeddings = word_embeddings + position_embeddings
+        else:
+            embeddings = word_embeddings
+
+        # Data format change to avoid explicit tranposes : [b s h] --> [s b h].
+        embeddings = embeddings.transpose(0, 1).contiguous()
+
+        if tokentype_ids is not None:
+            assert self.tokentype_embeddings is not None
+            # [b s h] -> [s b h] (So that it can be added with embeddings)
+            tokentype_embedding = self.tokentype_embeddings(tokentype_ids).permute(1, 0, 2)
+            embeddings = embeddings + tokentype_embedding
+        else:
+            assert self.tokentype_embeddings is None
+
+        # If the input flag for fp32 residual connection is set, convert for float.
+        if self.config.fp32_residual_connection:
+            embeddings = embeddings.float()
+
+        # Dropout.
+        if self.config.sequence_parallel:
+            embeddings = tensor_parallel.scatter_to_sequence_parallel_region(embeddings)
+            # `scatter_to_sequence_parallel_region` returns a view, which prevents
+            # the original tensor from being garbage collected. Clone to facilitate GC.
+            # Has a small runtime cost (~0.5%).
+            if self.config.clone_scatter_output_in_embedding:
+                embeddings = embeddings.clone()
+            with tensor_parallel.get_cuda_rng_tracker().fork():
+                embeddings = self.embedding_dropout(embeddings)
+        else:
+            embeddings = self.embedding_dropout(embeddings)
+
+        return embeddings

Megatron-LM-core_r0.7.0.beta/megatron/core/models/common/embeddings/rotary_pos_embedding.py

+# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Optional
+
+if TYPE_CHECKING:
+    from megatron.core.transformer.transformer_config import TransformerConfig
+    from megatron.core.transformer.transformer_block import TransformerBlock
+
+import logging
+
+import torch
+from torch import Tensor, nn
+
+from megatron.core import parallel_state
+
+logger = logging.getLogger(__name__)
+
+try:
+    from apex.transformer.functional import (
+        fused_apply_rotary_pos_emb,
+        fused_apply_rotary_pos_emb_thd,
+    )
+
+    HAVE_APPLY_ROPE_FUSION = True
+except:
+    HAVE_APPLY_ROPE_FUSION = False
+
+
+__all__ = ['RotaryEmbedding', 'apply_rotary_pos_emb']
+
+
+def get_pos_emb_on_this_cp_rank(pos_emb, seq_dim):
+    cp_size = parallel_state.get_context_parallel_world_size()
+    cp_rank = parallel_state.get_context_parallel_rank()
+    cp_idx = torch.tensor(
+        [cp_rank, (2 * cp_size - cp_rank - 1)], device="cpu", pin_memory=True
+    ).cuda(non_blocking=True)
+    pos_emb = pos_emb.view(
+        *pos_emb.shape[:seq_dim], 2 * cp_size, -1, *pos_emb.shape[(seq_dim + 1) :]
+    )
+    pos_emb = pos_emb.index_select(seq_dim, cp_idx)
+    pos_emb = pos_emb.view(*pos_emb.shape[:seq_dim], -1, *pos_emb.shape[(seq_dim + 2) :])
+    return pos_emb
+
+
+class RotaryEmbedding(nn.Module):
+    """Rotary Embedding for language model.
+
+    Args:
+        kv_channels (int): Projection weights dimension in multi-head attention. Obtained from transformer config
+        rotary_percent (float): Percent of rotary dimension to use for rotary position embeddings.
+        seq_len_interpolation_factor (float, optional): scale of linearly interpolating RoPE for longer sequences. The value must be a float larger than 1.0. Defaults to None
+        rotary_base (int, optional): Base period for rotary position embeddings. Defaults to 10000.
+    """
+
+    def __init__(
+        self,
+        kv_channels: int,
+        rotary_percent: float,
+        rotary_interleaved: bool = False,
+        seq_len_interpolation_factor: float = None,
+        rotary_base: int = 10000,
+    ) -> None:
+        super().__init__()
+
+        dim = kv_channels
+        if rotary_percent < 1.0:
+            dim = int(dim * rotary_percent)
+        self.rotary_interleaved = rotary_interleaved
+
+        self.seq_len_interpolation_factor = seq_len_interpolation_factor
+        self.inv_freq = 1.0 / (
+            rotary_base
+            ** (
+                torch.arange(0, dim, 2, dtype=torch.float32, device=torch.cuda.current_device())
+                / dim
+            )
+        )
+
+    def forward(self, max_seq_len: int, offset: int = 0) -> Tensor:
+        """Forward pass of RoPE embedding.
+
+        Args:
+            max_seq_len (int): Maximum size of sequence
+            offset (int, optional): _description_. Defaults to 0.
+
+        Returns:
+            Tensor: Embeddings after applying RoPE.
+        """
+        seq = (
+            torch.arange(max_seq_len, device=self.inv_freq.device, dtype=self.inv_freq.dtype)
+            + offset
+        )
+
+        if self.seq_len_interpolation_factor is not None:
+            seq *= 1 / self.seq_len_interpolation_factor
+
+        freqs = torch.outer(seq, self.inv_freq)
+        # first part even vector components, second part odd vector components,
+        #  2 * dim in dimension size
+        if not self.rotary_interleaved:
+            emb = torch.cat((freqs, freqs), dim=-1)
+        else:
+            emb = torch.stack((freqs.view(-1, 1), freqs.view(-1, 1)), dim=-1).view(
+                freqs.shape[0], -1
+            )
+        # emb [seq_length, .., dim]
+        emb = emb[:, None, None, :]
+        if parallel_state.get_context_parallel_world_size() > 1:
+            # slice rotary_pos_emb along sequence dimension and select the parition of the current CP rank
+            emb = get_pos_emb_on_this_cp_rank(emb, 0)
+        return emb
+
+    def _load_from_state_dict(self, state_dict, prefix, *args, **kwargs):
+        state_dict.pop(f'{prefix}inv_freq', None)
+        return super()._load_from_state_dict(state_dict, prefix, *args, **kwargs)
+
+    def get_rotary_seq_len(
+        self,
+        inference_params,
+        transformer: TransformerBlock,
+        transformer_input: Tensor,
+        transformer_config: TransformerConfig,
+    ) -> float:
+        """Function to get the rotary sequence length.
+
+        Args:
+            inference_params : Used during Inference time
+            transformer (TransformerBlock): The transformer block (decoder/encoder) used by the model
+            transformer_input (Tensor): _description_
+            transformer_config (TransformerConfig): Transformer config used by the model
+
+        Returns:
+            float: The rotary sequence length
+        """
+        if inference_params is not None:
+            rotary_seq_len = inference_params.max_sequence_length
+        else:
+            if transformer.input_tensor is not None:
+                rotary_seq_len = transformer.input_tensor.size(0)
+            else:
+                rotary_seq_len = transformer_input.size(0)
+
+            if transformer_config.sequence_parallel:
+                rotary_seq_len *= transformer_config.tensor_model_parallel_size
+
+        rotary_seq_len *= transformer_config.context_parallel_size
+
+        return rotary_seq_len
+
+
+def _rotate_half(x: Tensor, rotary_interleaved: bool) -> Tensor:
+    """Change sign so the last dimension becomes [-odd, +even]
+
+    Args:
+        x (Tensor): Input tensor
+
+    Returns:
+        Tensor: Tensor rotated half
+    """
+    if not rotary_interleaved:
+        x1, x2 = torch.chunk(x, 2, dim=-1)
+        return torch.cat((-x2, x1), dim=-1)
+    else:
+        x1 = x[:, :, :, ::2]
+        x2 = x[:, :, :, 1::2]
+        x_new = torch.stack((-x2, x1), dim=-1)
+        return x_new.view(x_new.shape[0], x_new.shape[1], x_new.shape[2], -1)
+
+
+def apply_rotary_pos_emb_bshd(t: Tensor, freqs: Tensor, rotary_interleaved: bool = False) -> Tensor:
+    """Apply rotary positional embedding to input tensor T.
+
+    check https://kexue.fm/archives/8265 for detailed formulas
+
+    Args:
+        t (Tensor): Input tensor T is of shape [seq_length, ... , dim]
+        freqs (Tensor): Rotary Positional embedding tensor freq is of shape [seq_length, ..., dim]
+
+    Returns:
+        Tensor: The input tensor after applying RoPE
+    """
+    rot_dim = freqs.shape[-1]
+
+    # ideally t_pass is empty so rotary pos embedding is applied to all tensor t
+    t, t_pass = t[..., :rot_dim], t[..., rot_dim:]
+
+    # first part is cosine component
+    # second part is sine component, need to change signs with _rotate_half method
+    cos_ = torch.cos(freqs).to(t.dtype)
+    sin_ = torch.sin(freqs).to(t.dtype)
+
+    t = (t * cos_) + (_rotate_half(t, rotary_interleaved) * sin_)
+    return torch.cat((t, t_pass), dim=-1)
+
+
+def apply_rotary_pos_emb_thd(
+    t: Tensor, cu_seqlens: Tensor, freqs: Tensor, rotary_interleaved: bool = False
+) -> Tensor:
+
+    """A baseline implementation of applying RoPE for `thd` format.
+
+    Args:
+        t (Tensor): Input tensor T is of shape [t, h, d]
+        cu_seqlens(Tensor):  Cumulative sum of sequence lengths in a batch for `t`,
+        with shape [b + 1] and dtype torch.int32.
+        freqs (Tensor): Rotary Positional embedding tensor freq is of shape [max_s, 1, 1, d]
+
+    Returns:
+        Tensor: Shape [t, h, d]. The input tensor after applying RoPE.
+    """
+
+    seqlens = (cu_seqlens[1:] - cu_seqlens[:-1]).tolist()
+    return torch.cat(
+        [
+            apply_rotary_pos_emb_bshd(x.unsqueeze(1), freqs[: x.size(0)])
+            for x in torch.split(t, seqlens)
+        ]
+    ).squeeze(1)
+
+
+def apply_rotary_pos_emb(
+    t: Tensor, freqs: Tensor, config: TransformerConfig, cu_seqlens: Optional[Tensor] = None,
+):
+    """
+    Reroute to the appropriate apply_rotary_pos_emb function depending on
+    fused/unfused kernels, or bshd (conventional) / thd (packed seq) format
+    """
+    if config.apply_rope_fusion and not HAVE_APPLY_ROPE_FUSION:
+        # setting apply_rope_fusion in config to False so that subsequent queries to this config also return False
+        config.apply_rope_fusion = False
+        if not getattr(apply_rotary_pos_emb, "printed_fused_warning", False):
+            logger.warning(
+                "Setting apply_rope_fusion to false because its implementation"
+                " is not included in Apex. Try upgrading to the latest version"
+            )
+            apply_rotary_pos_emb.printed_fused_warning = True
+    if config.apply_rope_fusion:
+        if cu_seqlens is None:
+            return fused_apply_rotary_pos_emb(t, freqs, transpose_output_memory=True)
+        else:
+            return fused_apply_rotary_pos_emb_thd(t, cu_seqlens, freqs)
+    else:
+        if cu_seqlens is None:
+            return apply_rotary_pos_emb_bshd(t, freqs, rotary_interleaved=config.rotary_interleaved)
+        else:
+            return apply_rotary_pos_emb_thd(
+                t, cu_seqlens, freqs, rotary_interleaved=config.rotary_interleaved
+            )

Megatron-LM-core_r0.7.0.beta/megatron/core/models/common/language_module/language_module.py

+import logging
+from typing import Optional, Tuple
+
+import torch
+from torch import Tensor
+
+from megatron.core import parallel_state, tensor_parallel
+from megatron.core.dist_checkpointing.mapping import ShardedStateDict
+from megatron.core.transformer.module import MegatronModule
+from megatron.core.transformer.transformer_config import TransformerConfig
+from megatron.core.utils import make_tp_sharded_tensor_for_checkpoint
+
+
+class LanguageModule(MegatronModule):
+    """Base language module that has common helper functions used across GPT, BERT etc.
+
+    Args:
+        config (TransformerConfig): Input transformer config for the model
+    """
+
+    def __init__(self, config: TransformerConfig) -> None:
+        super().__init__(config=config)
+
+    def compute_language_model_loss(self, labels: Tensor, logits: Tensor) -> Tensor:
+        """Computes the language model loss (Cross entropy across vocabulary)
+
+        Args:
+            labels (Tensor): The labels of dimension [batch size, seq length]
+            logits (Tensor): The final logits returned by the output layer of the transformer model
+
+        Returns:
+            Tensor: Loss tensor of dimensions [batch size, sequence_length]
+        """
+        # [b s] => [s b]
+        labels = labels.transpose(0, 1).contiguous()
+        loss = tensor_parallel.vocab_parallel_cross_entropy(logits.float(), labels)
+
+        # [s b] => [b, s]
+        loss = loss.transpose(0, 1).contiguous()
+        return loss
+
+    def setup_embeddings_and_output_layer(self) -> None:
+        """Sets up embedding layer in first stage and output layer in last stage.
+
+        This function initalizes word embeddings in the final stage when we are
+        using pipeline parallelism and sharing word embeddings, and sets up param
+        attributes on the embedding and output layers.
+        """
+
+        # Set `is_embedding_or_output_parameter` attribute.
+        if self.pre_process:
+            self.embedding.word_embeddings.weight.is_embedding_or_output_parameter = True
+        if self.post_process and self.output_layer.weight is not None:
+            self.output_layer.weight.is_embedding_or_output_parameter = True
+
+        if not self.share_embeddings_and_output_weights:
+            return
+
+        if self.pre_process and self.post_process:
+            # Zero out wgrad if sharing embeddings between two layers on same
+            # pipeline stage to make sure grad accumulation into main_grad is
+            # correct and does not include garbage values (e.g., from torch.empty).
+            self.shared_embedding_or_output_weight().zero_out_wgrad = True
+            return
+
+        if self.pre_process and not self.post_process:
+            assert parallel_state.is_pipeline_first_stage()
+            self.shared_embedding_or_output_weight().shared_embedding = True
+
+        if self.post_process and not self.pre_process:
+            assert not parallel_state.is_pipeline_first_stage()
+            # set word_embeddings weights to 0 here, then copy first
+            # stage's weights using all_reduce below.
+            self.output_layer.weight.data.fill_(0)
+            self.output_layer.weight.shared = True
+            self.output_layer.weight.shared_embedding = True
+
+        # Parameters are shared between the word embeddings layers, and the
+        # heads at the end of the model. In a pipelined setup with more than
+        # one stage, the initial embedding layer and the head are on different
+        # workers, so we do the following:
+        # 1. Create a second copy of word_embeddings on the last stage, with
+        #    initial parameters of 0.0.
+        # 2. Do an all-reduce between the first and last stage to ensure that
+        #    the two copies of word_embeddings start off with the same
+        #    parameter values.
+        # 3. In the training loop, before an all-reduce between the grads of
+        #    the two word_embeddings layers to ensure that every applied weight
+        #    update is the same on both stages.
+
+        # Ensure that first and last stages have the same initial parameter
+        # values.
+        if torch.distributed.is_initialized():
+            if parallel_state.is_rank_in_embedding_group():
+                weight = self.shared_embedding_or_output_weight()
+                weight.data = weight.data.cuda()
+                torch.distributed.all_reduce(
+                    weight.data, group=parallel_state.get_embedding_group()
+                )
+
+        elif not getattr(LanguageModule, "embedding_warning_printed", False):
+            logging.getLogger(__name__).warning(
+                "Distributed processes aren't initialized, so the output layer "
+                "is not initialized with weights from the word embeddings. "
+                "If you are just manipulating a model this is fine, but "
+                "this needs to be handled manually. If you are training "
+                "something is definitely wrong."
+            )
+            LanguageModule.embedding_warning_printed = True
+
+    def shared_embedding_or_output_weight(self) -> Tensor:
+        """Gets the emedding weight or output logit weights when share embedding and output weights set to True.
+
+        Returns:
+            Tensor: During pre processing it returns the input embeddings weight while during post processing it returns the final output layers weight
+        """
+        if self.pre_process:
+            return self.embedding.word_embeddings.weight
+        elif self.post_process:
+            return self.output_layer.weight
+        return None
+
+    def sharded_state_dict(
+        self,
+        prefix: str = '',
+        sharded_offsets: Tuple[Tuple[int, int, int]] = (),
+        metadata: Optional[dict] = None,
+    ) -> ShardedStateDict:
+        """ Sharded state dict implementation that handles the output layer weights tying.
+
+        Args:
+            prefix (str): Module name prefix.
+            sharded_offsets (tuple): PP related offsets, expected to be empty at this module level.
+            metadata (Optional[Dict]): metadata controlling sharded state dict creation.
+
+        Returns:
+            ShardedStateDict: sharded state dict for the LanguageModel
+        """
+        assert not sharded_offsets, "Unexpected sharded offsets"
+        sharded_state_dict = super().sharded_state_dict(prefix, sharded_offsets, metadata)
+
+        first_stage_word_emb_key = f'{prefix}embedding.word_embeddings.weight'
+        output_layer_weight_key = f'{prefix}output_layer.weight'
+        output_layer_bias_key = f'{prefix}output_layer.bias'
+
+        if self.share_embeddings_and_output_weights:
+            self.tie_embeddings_and_output_weights_state_dict(
+                sharded_state_dict, output_layer_weight_key, first_stage_word_emb_key
+            )
+        elif self.post_process:
+            # Make sure the output layer follows the embeddings padding logic
+            sharded_state_dict[output_layer_weight_key].allow_shape_mismatch = True
+
+        # Regardless of sharing the output weights with embeddings, we must handle the bias padding
+        if self.post_process and output_layer_bias_key in sharded_state_dict:
+            sharded_state_dict[output_layer_bias_key].allow_shape_mismatch = True
+
+        return sharded_state_dict
+
+    def tie_embeddings_and_output_weights_state_dict(
+        self,
+        sharded_state_dict: ShardedStateDict,
+        output_layer_weight_key: str,
+        first_stage_word_emb_key: str,
+    ) -> None:
+        """Ties the embedding and output weights in a given sharded state dict.
+
+        Args:
+            sharded_state_dict (ShardedStateDict): state dict with the weight to tie
+            output_layer_weight_key (str): key of the output layer weight in the state dict.
+                This entry will be replaced with a tied version
+            first_stage_word_emb_key (str): this must be the same as the
+                ShardedTensor.key of the first stage word embeddings.
+
+        Returns: None, acts in-place
+        """
+        if not self.post_process:
+            # No output layer
+            assert output_layer_weight_key not in sharded_state_dict, sharded_state_dict.keys()
+            return
+
+        if self.pre_process:
+            # Output layer is equivalent to the embedding already
+            return
+
+        # Replace the default output layer with a one sharing the weights with the embedding
+        del sharded_state_dict[output_layer_weight_key]
+        tensor = self.shared_embedding_or_output_weight()
+        last_stage_word_emb_replica_id = (
+            1,  # copy of first stage embedding
+            0,
+            parallel_state.get_data_parallel_rank(with_context_parallel=True),
+        )
+
+        sharded_state_dict[output_layer_weight_key] = make_tp_sharded_tensor_for_checkpoint(
+            tensor=tensor,
+            key=first_stage_word_emb_key,
+            replica_id=last_stage_word_emb_replica_id,
+            allow_shape_mismatch=True,
+        )

Megatron-LM-core_r0.7.0.beta/megatron/core/models/common/vision_module/vision_module.py

+# Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
+"""Megatron Vision Module."""
+
+from megatron.core.transformer.module import MegatronModule
+from megatron.core.transformer.transformer_config import TransformerConfig
+
+
+# Note: This is only a stub at the moment. This will be expanded in follow-up changes.
+class VisionModule(MegatronModule):
+    """Base vision module that has common helper functions used across CLIP, ViT, etc.
+
+    Args:
+        config (TransformerConfig): Input transformer config for the model
+    """
+
+    def __init__(self, config: TransformerConfig) -> None:
+        super().__init__(config=config)

Megatron-LM-core_r0.7.0.beta/megatron/core/models/gpt/__init__.py

+from .gpt_model import GPTModel

Megatron-LM-core_r0.7.0.beta/megatron/core/models/gpt/gpt_layer_specs.py

+# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
+
+from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
+from megatron.core.fusions.fused_layer_norm import FusedLayerNorm
+from megatron.core.tensor_parallel.layers import ColumnParallelLinear, RowParallelLinear
+from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
+from megatron.core.transformer.custom_layers.transformer_engine import (
+    TEDotProductAttention,
+    TELayerNormColumnParallelLinear,
+    TENorm,
+    TERowParallelLinear,
+)
+from megatron.core.transformer.dot_product_attention import DotProductAttention
+from megatron.core.transformer.enums import AttnMaskType
+from megatron.core.transformer.identity_op import IdentityOp
+from megatron.core.transformer.mlp import MLP, MLPSubmodules
+from megatron.core.transformer.moe.moe_layer import MoELayer
+from megatron.core.transformer.spec_utils import ModuleSpec
+from megatron.core.transformer.transformer_block import TransformerBlockSubmodules
+from megatron.core.transformer.transformer_layer import TransformerLayer, TransformerLayerSubmodules
+
+
+# Use this spec to use lower level Transformer Engine modules (required for fp8 training)
+def get_gpt_layer_with_transformer_engine_spec(
+    num_experts: int = None, moe_grouped_gemm: bool = False, qk_layernorm: bool = False
+) -> ModuleSpec:
+    mlp = _get_mlp_module_spec(
+        use_te=True, num_experts=num_experts, moe_grouped_gemm=moe_grouped_gemm
+    )
+    return ModuleSpec(
+        module=TransformerLayer,
+        submodules=TransformerLayerSubmodules(
+            self_attention=ModuleSpec(
+                module=SelfAttention,
+                params={"attn_mask_type": AttnMaskType.causal},
+                submodules=SelfAttentionSubmodules(
+                    linear_qkv=TELayerNormColumnParallelLinear,
+                    core_attention=TEDotProductAttention,
+                    linear_proj=TERowParallelLinear,
+                    q_layernorm=TENorm if qk_layernorm else IdentityOp,
+                    k_layernorm=TENorm if qk_layernorm else IdentityOp,
+                ),
+            ),
+            self_attn_bda=get_bias_dropout_add,
+            pre_mlp_layernorm=TENorm if num_experts else IdentityOp,
+            mlp=mlp,
+            mlp_bda=get_bias_dropout_add,
+        ),
+    )
+
+
+# Use this spec for an implementation using only modules in megatron core
+def get_gpt_layer_local_spec(
+    num_experts: int = None, moe_grouped_gemm: bool = False, qk_layernorm: bool = False
+) -> ModuleSpec:
+    mlp = _get_mlp_module_spec(
+        use_te=False, num_experts=num_experts, moe_grouped_gemm=moe_grouped_gemm
+    )
+    return ModuleSpec(
+        module=TransformerLayer,
+        submodules=TransformerLayerSubmodules(
+            input_layernorm=FusedLayerNorm,
+            self_attention=ModuleSpec(
+                module=SelfAttention,
+                params={"attn_mask_type": AttnMaskType.causal},
+                submodules=SelfAttentionSubmodules(
+                    linear_qkv=ColumnParallelLinear,
+                    core_attention=DotProductAttention,
+                    linear_proj=RowParallelLinear,
+                    q_layernorm=FusedLayerNorm if qk_layernorm else IdentityOp,
+                    k_layernorm=FusedLayerNorm if qk_layernorm else IdentityOp,
+                ),
+            ),
+            self_attn_bda=get_bias_dropout_add,
+            pre_mlp_layernorm=FusedLayerNorm,
+            mlp=mlp,
+            mlp_bda=get_bias_dropout_add,
+            sharded_state_dict_keys_map={
+                'input_layernorm.': 'self_attention.linear_qkv.layer_norm_',
+                'pre_mlp_layernorm.': 'mlp.linear_fc1.layer_norm_',
+            },
+        ),
+    )
+
+
+# Helper function to get module spec for MLP/MoE
+def _get_mlp_module_spec(
+    use_te: bool = True, num_experts: int = None, moe_grouped_gemm: bool = False
+) -> ModuleSpec:
+    if num_experts is None:
+        # Dense MLP w/ or w/o TE modules.
+        return ModuleSpec(
+            module=MLP,
+            submodules=MLPSubmodules(
+                linear_fc1=TELayerNormColumnParallelLinear if use_te else ColumnParallelLinear,
+                linear_fc2=TERowParallelLinear if use_te else RowParallelLinear,
+            ),
+        )
+    else:
+        # Mixture of experts with modules in megatron core.
+        return ModuleSpec(
+            module=MoELayer,
+            submodules=MLPSubmodules(linear_fc1=ColumnParallelLinear, linear_fc2=RowParallelLinear,)
+            if not moe_grouped_gemm
+            else None,
+        )

Megatron-LM-core_r0.7.0.beta/megatron/core/models/gpt/gpt_model.py

+# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
+
+import logging
+from typing import Dict, Literal, Optional, Tuple, Union
+
+import torch
+from torch import Tensor
+
+from megatron.core import InferenceParams, parallel_state, tensor_parallel
+from megatron.core.dist_checkpointing.mapping import ShardedStateDict
+from megatron.core.models.common.embeddings.language_model_embedding import LanguageModelEmbedding
+from megatron.core.models.common.embeddings.rotary_pos_embedding import RotaryEmbedding
+from megatron.core.models.common.language_module.language_module import LanguageModule
+from megatron.core.packed_seq_params import PackedSeqParams
+from megatron.core.transformer.enums import AttnMaskType, ModelType
+from megatron.core.transformer.spec_utils import ModuleSpec
+from megatron.core.transformer.transformer_block import TransformerBlock
+from megatron.core.transformer.transformer_config import TransformerConfig
+from megatron.core.utils import make_tp_sharded_tensor_for_checkpoint
+
+
+class GPTModel(LanguageModule):
+    """GPT Transformer language model.
+
+    Args:
+        config (TransformerConfig): Transformer config
+        transformer_layer_spec (ModuleSpec): Specifies module to use for transformer layers
+        vocab_size (int): Vocabulary size
+        max_sequence_length (int): maximum size of sequence. This is used for positional embedding
+        pre_process (bool, optional): Include embedding layer (used with pipeline parallelism). Defaults to True.
+        post_process (bool, optional): Include an output layer (used with pipeline parallelism). Defaults to True.
+        fp16_lm_cross_entropy (bool, optional): Defaults to False.
+        parallel_output (bool, optional): Do not gather the outputs, keep them split across tensor parallel ranks. Defaults to True.
+        share_embeddings_and_output_weights (bool, optional): When True, input embeddings and output logit weights are shared. Defaults to False.
+        position_embedding_type (Literal[learned_absolute,rope], optional):  Position embedding type.. Defaults to 'learned_absolute'.
+        rotary_percent (float, optional): Percent of rotary dimension to use for rotary position embeddings. Ignored unless position_embedding_type is 'rope'. Defaults to 1.0.
+        rotary_base (int, optional): Base period for rotary position embeddings. Ignored unless position_embedding_type is 'rope'. Defaults to 10000.
+        seq_len_interpolation_factor (Optional[float], optional): scale of linearly interpolating RoPE for longer sequences. The value must be a float larger than 1.0. Defaults to None.
+    """
+
+    def __init__(
+        self,
+        config: TransformerConfig,
+        transformer_layer_spec: ModuleSpec,
+        vocab_size: int,
+        max_sequence_length: int,
+        pre_process: bool = True,
+        post_process: bool = True,
+        fp16_lm_cross_entropy: bool = False,
+        parallel_output: bool = True,
+        share_embeddings_and_output_weights: bool = False,
+        position_embedding_type: Literal['learned_absolute', 'rope'] = 'learned_absolute',
+        rotary_percent: float = 1.0,
+        rotary_base: int = 10000,
+        seq_len_interpolation_factor: Optional[float] = None,
+    ) -> None:
+        super().__init__(config=config)
+
+        self.transformer_layer_spec: ModuleSpec = transformer_layer_spec
+        self.vocab_size = vocab_size
+        self.max_sequence_length = max_sequence_length
+        self.pre_process = pre_process
+        self.post_process = post_process
+        self.fp16_lm_cross_entropy = fp16_lm_cross_entropy
+        self.parallel_output = parallel_output
+        self.share_embeddings_and_output_weights = share_embeddings_and_output_weights
+        self.position_embedding_type = position_embedding_type
+
+        # megatron core pipelining currently depends on model type
+        # TODO: remove this dependency ?
+        self.model_type = ModelType.encoder_or_decoder
+
+        # These 2 attributes are needed for TensorRT-LLM export.
+        self.max_position_embeddings = max_sequence_length
+        self.rotary_percent = rotary_percent
+
+        if self.pre_process:
+            self.embedding = LanguageModelEmbedding(
+                config=self.config,
+                vocab_size=self.vocab_size,
+                max_sequence_length=self.max_sequence_length,
+                position_embedding_type=position_embedding_type,
+            )
+
+        if self.position_embedding_type == 'rope':
+            self.rotary_pos_emb = RotaryEmbedding(
+                kv_channels=self.config.kv_channels,
+                rotary_percent=rotary_percent,
+                rotary_interleaved=self.config.rotary_interleaved,
+                seq_len_interpolation_factor=seq_len_interpolation_factor,
+                rotary_base=rotary_base,
+            )
+
+        # Transformer.
+        self.decoder = TransformerBlock(
+            config=self.config,
+            spec=transformer_layer_spec,
+            pre_process=self.pre_process,
+            post_process=self.post_process,
+        )
+
+        # Output
+        if post_process:
+            if self.config.defer_embedding_wgrad_compute:
+                # The embedding activation buffer preserves a reference to the input activations
+                # of the final embedding projection layer GEMM. It will hold the activations for
+                # all the micro-batches of a global batch for the last pipeline stage. Once we are
+                # done with all the back props for all the microbatches for the last pipeline stage,
+                # it will be in the pipeline flush stage. During this pipeline flush we use the
+                # input activations stored in embedding activation buffer and gradient outputs stored
+                # in gradient buffer to calculate the weight gradients for the embedding final linear layer.
+                self.embedding_activation_buffer = []
+                self.grad_output_buffer = []
+            else:
+                self.embedding_activation_buffer = None
+                self.grad_output_buffer = None
+
+            self.output_layer = tensor_parallel.ColumnParallelLinear(
+                config.hidden_size,
+                self.vocab_size,
+                config=config,
+                init_method=config.init_method,
+                bias=False,
+                skip_bias_add=False,
+                gather_output=not self.parallel_output,
+                skip_weight_param_allocation=self.pre_process
+                and self.share_embeddings_and_output_weights,
+                embedding_activation_buffer=self.embedding_activation_buffer,
+                grad_output_buffer=self.grad_output_buffer,
+            )
+
+        if self.pre_process or self.post_process:
+            self.setup_embeddings_and_output_layer()
+
+    def set_input_tensor(self, input_tensor: Tensor) -> None:
+        """Sets input tensor to the model.
+
+        See megatron.model.transformer.set_input_tensor()
+
+        Args:
+            input_tensor (Tensor): Sets the input tensor for the model.
+        """
+        # This is usually handled in schedules.py but some inference code still
+        # gives us non-lists or None
+        if not isinstance(input_tensor, list):
+            input_tensor = [input_tensor]
+
+        assert len(input_tensor) == 1, 'input_tensor should only be length 1 for gpt/bert'
+        self.decoder.set_input_tensor(input_tensor[0])
+
+    def forward(
+        self,
+        input_ids: Tensor,
+        position_ids: Tensor,
+        attention_mask: Tensor,
+        decoder_input: Tensor = None,
+        labels: Tensor = None,
+        inference_params: InferenceParams = None,
+        packed_seq_params: PackedSeqParams = None,
+        extra_block_kwargs: dict = None,
+    ) -> Tensor:
+        """Forward function of the GPT Model This function passes the input tensors
+        through the embedding layer, and then the decoeder and finally into the post
+        processing layer (optional).
+
+        It either returns the Loss values if labels are given  or the final hidden units
+        """
+        # If decoder_input is provided (not None), then input_ids and position_ids are ignored.
+        # Otherwise, apply embedding layer on input_ids and position_ids to get decoder_input.
+
+        # Decoder embedding.
+        if decoder_input is not None:
+            pass
+        elif self.pre_process:
+            decoder_input = self.embedding(input_ids=input_ids, position_ids=position_ids)
+        else:
+            # intermediate stage of pipeline
+            # decoder will get hidden_states from encoder.input_tensor
+            decoder_input = None
+
+        # Rotary positional embeddings (embedding is None for PP intermediate devices)
+        rotary_pos_emb = None
+        if self.position_embedding_type == 'rope':
+            rotary_seq_len = self.rotary_pos_emb.get_rotary_seq_len(
+                inference_params, self.decoder, decoder_input, self.config
+            )
+            rotary_pos_emb = self.rotary_pos_emb(rotary_seq_len)
+
+        # Run decoder.
+        hidden_states = self.decoder(
+            hidden_states=decoder_input,
+            attention_mask=attention_mask,
+            inference_params=inference_params,
+            rotary_pos_emb=rotary_pos_emb,
+            packed_seq_params=packed_seq_params,
+            **(extra_block_kwargs or {}),
+        )
+
+        if not self.post_process:
+            return hidden_states
+
+        # logits and loss
+        output_weight = None
+        if self.share_embeddings_and_output_weights:
+            output_weight = self.shared_embedding_or_output_weight()
+        logits, _ = self.output_layer(hidden_states, weight=output_weight)
+
+        if labels is None:
+            # [s b h] => [b s h]
+            return logits.transpose(0, 1).contiguous()
+
+        loss = self.compute_language_model_loss(labels, logits)
+
+        return loss
+
+    def sharded_state_dict(
+        self, prefix: str = '', sharded_offsets: tuple = (), metadata: Optional[Dict] = None
+    ) -> ShardedStateDict:
+        """ Sharded state dict implementation for GPTModel backward-compatibility (removing extra state).
+
+        Args:
+            prefix (str): Module name prefix.
+            sharded_offsets (tuple): PP related offsets, expected to be empty at this module level.
+            metadata (Optional[Dict]): metadata controlling sharded state dict creation.
+
+        Returns:
+            ShardedStateDict: sharded state dict for the GPTModel
+        """
+        sharded_state_dict = super().sharded_state_dict(prefix, sharded_offsets, metadata)
+        output_layer_extra_state_key = f'{prefix}output_layer._extra_state'
+
+        # Old GPT checkpoints only stored the output layer weight key. So we remove the _extra_state key
+        # but check that it doesn't contain any data anyway
+        output_extra_state = sharded_state_dict.pop(output_layer_extra_state_key, None)
+        assert not (
+            output_extra_state and output_extra_state.data
+        ), f'Expected output layer extra state to be empty, got: {output_extra_state}'
+
+        return sharded_state_dict

Megatron-LM-core_r0.7.0.beta/megatron/core/models/multimodal/llava_model.py

+# Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
+import logging
+from collections import namedtuple
+from functools import partial
+from typing import List
+
+import torch
+
+from megatron.core import parallel_state, tensor_parallel
+from megatron.core.models.gpt import GPTModel
+from megatron.core.models.vision.clip_vit_model import CLIPViTModel
+from megatron.core.models.vision.multimodal_projector import MultimodalProjector
+from megatron.core.transformer import MegatronModule
+from megatron.core.transformer.spec_utils import ModuleSpec
+from megatron.core.transformer.transformer_config import TransformerConfig
+
+
+# Note: This is under development and may be missing features.
+class LLaVAModel(MegatronModule):
+    """LLaVA multi-modal model.
+
+    Args:
+        language_transformer_config (TransformerConfig): Transformer config for the language model.
+        language_transformer_layer_spec (ModuleSpec): Specifies module to use for transformer layers of the language model.
+        vocab_size (int): Vocabulary size.
+        max_sequence_length (int): maximum sequence length. This is used for positional embedding.
+        vision_transformer_config (TransformerConfig): Transformer config for the vision model.
+        vision_transformer_layer_spec (ModuleSpec): Specifies module to use for transformer layers of the vision model.
+        vision_projection_config (TransformerConfig): Config for the projection from vision model outputs to language model inputs.
+        vision_projection_layer_spec (ModuleSpec): Specifies the module to use for the vision projection.
+        vision_projection_type (str): Type of the vision projection to use. Default is a 2-layer MLP.
+        allow_missing_vision_projection_checkpoint (bool): Allow vision projection weights to be missing when loading a checkpoint. Default False.
+    """
+
+    def __init__(
+        self,
+        language_transformer_config: TransformerConfig,
+        language_transformer_layer_spec: ModuleSpec,
+        vocab_size: int,
+        max_sequence_length: int,
+        vision_transformer_config: TransformerConfig,
+        vision_transformer_layer_spec: ModuleSpec,
+        vision_projection_config: TransformerConfig,
+        vision_projection_layer_spec: ModuleSpec,
+        vision_projection_type: str = "mlp",
+        allow_missing_vision_projection_checkpoint: bool = False,
+    ) -> None:
+        super().__init__(config=language_transformer_config)
+
+        logging.getLogger(__name__).warning(
+            "LLaVA model is under development and may be missing features."
+        )
+
+        if parallel_state.get_pipeline_model_parallel_world_size() > 1:
+            raise NotImplementedError("pipeline parallelism is not supported in this model yet.")
+
+        self.language_model = GPTModel(
+            language_transformer_config,
+            language_transformer_layer_spec,
+            vocab_size,
+            max_sequence_length,
+        )
+
+        self.vision_model = CLIPViTModel(vision_transformer_config, vision_transformer_layer_spec)
+
+        # Map (intermediate) vision model outputs to the language model input dimension.
+        self.vision_projection = MultimodalProjector(
+            vision_projection_config,
+            vision_projection_layer_spec,
+            vision_projection_type,
+            vision_transformer_config.hidden_size,  # input size to the projection.
+        )
+
+        # This allows ignoring missing weights for the vision projection during checkpoint loading.
+        # This should be disabled by default but can be enabled if your checkpoint contains pretrained
+        # vision and language models but not the projection from vision model outputs to language model inputs.
+        if allow_missing_vision_projection_checkpoint:
+            vision_projection_param_names = [
+                f"vision_projection.{name}" for name in self.vision_projection.state_dict().keys()
+            ]
+            self.vision_projection.register_load_state_dict_post_hook(
+                partial(_load_state_dict_hook_ignore_param_names, vision_projection_param_names)
+            )
+
+    def set_input_tensor(self, input_tensor: torch.Tensor) -> None:
+        """Sets input tensor to the model.
+
+        NOTE: Pipeline parallelism is not supported in this model yet. This is just a placeholder implementation.
+
+        Args:
+            input_tensor (Tensor): Sets the input tensor for the model.
+        """
+        self.vision_model.set_input_tensor(input_tensor)
+
+    def freeze(
+        self, freeze_language_model: bool, freeze_vision_model: bool, freeze_vision_projection: bool
+    ):
+        """Freeze model modules.
+
+        Make specific modules non-trainable by setting requires_grad to False for the module's parameters.
+
+        Args:
+            freeze_language_model (bool): Freeze the language model module.
+            freeze_vision_model (bool): Freeze the vision model module.
+            freeze_vision_projection (bool): Freeze the vision projection module.
+        """
+        modules = []
+        if freeze_language_model:
+            modules.append(self.language_model)
+        if freeze_vision_model:
+            modules.append(self.vision_model)
+        if freeze_vision_projection:
+            modules.append(self.vision_projection)
+
+        for module in modules:
+            for param in module.parameters():
+                param.requires_grad = False
+
+    def forward(
+        self,
+        images: torch.Tensor,
+        input_ids: torch.Tensor,
+        position_ids: torch.Tensor,
+        attention_mask: torch.Tensor,
+        labels: torch.Tensor = None,
+    ) -> torch.Tensor:
+        """Forward function of the LLaVA model.
+
+        Args:
+            images (torch.Tensor): input image of shape [batch, img_h, img_w].
+            input_ids (torch.Tensor): input text ids [batch, text_seq_len].
+            position_ids (torch.Tensor): input text position ids [batch, text_seq_len].
+            attention_mask (torch.Tensor): attention mask for the language model [batch, 1, combined_seq_len, combined_seq_len].
+            labels (torch.Tensor): Optional target text labels [batch, combined_seq_len].
+
+        Returns:
+            output (torch.Tensor): Loss of shape [b, s] if labels are provided, otherwise logits of shape [b, s, vocab_size].
+        """
+        image_embeddings = self.vision_model(images)  # [b, img_seq_len, h_vision]
+
+        # map vision model output size to language model input size.
+        image_embeddings = self.vision_projection(image_embeddings)  # [b, img_seq_len, h_language]
+
+        image_embeddings = image_embeddings.permute(1, 0, 2)  # [img_seq_len, b, h_language]
+        language_embeddings = self.language_model.embedding(
+            input_ids=input_ids, position_ids=position_ids
+        )  # [text_seq_len, b, h_language]
+        combined_embeddings = torch.cat(
+            [image_embeddings, language_embeddings], dim=0
+        )  # [combined_seq_len, b, h_language]
+
+        # Embedding is computed above so we can discard input and position ids.
+        input_ids = None
+        position_ids = None
+
+        # Note: This returns loss if labels are provided, otherwise logits.
+        output = self.language_model(
+            input_ids,
+            position_ids,
+            attention_mask,
+            decoder_input=combined_embeddings,
+            labels=labels,
+        )
+
+        return output
+
+
+def _load_state_dict_hook_ignore_param_names(
+    param_names: List[str], module: torch.nn.Module, incompatible_keys: namedtuple
+):
+    """Hook to ignore missing keys during checkpoint loading.
+
+    By default, this should not be used to avoid accidentally missing weights in checkpoint loading.
+
+    Example use case: Use this for the vision projection if you want to load a checkpoint that contains vision and language model weights
+    but not the vision projection weights.
+
+    Args:
+        param_names (list of str): Parameter names allowed to be missing when calling load_state_dict.
+        module (torch.nn.Module): The torch module this hook applies to. Unused here but required by the torch API.
+        incompatible_keys (namedtuple): Namedtuple with fields missing_keys and unexpected_keys, which collect the missing and unexpected
+            keys when calling load_state_dict on this torch module, respectively.
+    """
+    for param_name in param_names:
+        incompatible_keys.missing_keys.remove(param_name)

Megatron-LM-core_r0.7.0.beta/megatron/core/models/retro/__init__.py

+# Copyright (c) 2024, NVIDIA CORPORATION.  All rights reserved.
+
+"""
+Exports:
+
+  - RetroConfig: configuration dataclass for RetroModel.
+  - RetroModel: The Retro model.
+  - get_retro_decoder_block_spec: Get spec for Retro decoder transformer block.
+"""
+
+from .config import RetroConfig
+from .decoder_spec import get_retro_decoder_block_spec
+from .model import RetroModel

Megatron-LM-core_r0.7.0.beta/megatron/core/models/retro/base_attention.py

+# Copyright (c) 2024, NVIDIA CORPORATION.  All rights reserved.
+
+"""Base class for decoder and encoder attention modules."""
+
+from megatron.core.models.retro.config import RetroConfig
+from megatron.core.transformer.attention import CrossAttention, CrossAttentionSubmodules
+from megatron.core.transformer.enums import AttnMaskType
+from megatron.core.transformer.module import MegatronModule
+
+
+class BaseRetroCrossAttention(MegatronModule):
+
+    """Base class for Retro cross attention, for both encoder & decoder layers.
+
+    This class collects the retro arguments below (i.e., num neighbors, chunk
+    length, and retrieve length) for use in Retro's custom cross attention
+    operators.
+
+    Args:
+        config (RetroConfig): Retro config.
+        submodules (CrossAttentionSubmodules): Cross attention submodules.
+        layer_number (int): Layer number within transformer block.
+        attn_mask_type (AttnMaskType): Mask type ('causal' or 'padding').
+    """
+
+    def __init__(
+        self,
+        config: RetroConfig,
+        submodules: CrossAttentionSubmodules,
+        layer_number: int = 1,
+        attn_mask_type: AttnMaskType = AttnMaskType.padding,
+    ):
+        super().__init__(config=config)
+
+        self.attn = CrossAttention(
+            config=config,
+            submodules=submodules,
+            layer_number=layer_number,
+            attn_mask_type=attn_mask_type,
+        )
+
+        self.retro_num_neighbors = config.retro_num_neighbors
+        self.retro_chunk_length = config.retro_chunk_length
+        self.retro_retrieved_length = config.retro_retrieved_length

Megatron-LM-core_r0.7.0.beta/megatron/core/models/retro/config.py

+# Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
+
+"""Configuration dataclass for a RetroModel."""
+
+import os
+import types
+from dataclasses import dataclass
+from importlib.metadata import version
+
+from pkg_resources import packaging
+
+from megatron.core.transformer import TransformerConfig
+
+
+@dataclass
+class RetroConfig(TransformerConfig):
+    """Configuration object for Retro models. """
+
+    # Retro.
+    retro_project_dir: str = None
+    """Retro project directory, which contains the preprocessed data for for pretraining. This
+       directory is built during preprocessing (see tools/retro/README.md), and contains
+       subdirectories for the chunk database and pretraining neighbors.
+    """
+
+    retro_block_size: int = None
+    """Number of records to load per data file, as saved during preprocessing. Block processing is
+       used for efficient data preprocessing.
+    """
+
+    retro_chunk_length: int = None
+    """Chunk length used for performing chunked- cross-attention (CCA)."""
+
+    retro_encoder_num_layers: int = 2
+    """Number of layers to use for the retrieval encoder."""
+
+    retro_encoder_hidden_dropout: float = 0.1
+    """Hidden dropout for retrieval encoder."""
+
+    retro_encoder_attention_dropout: float = 0.1
+    """Attention dropout for retrieval encoder."""
+
+    retro_neighbor_dirs: dict = None
+    """Directory names of saved neighbor id files for train, valid, and test datasets."""
+
+    retro_num_neighbors: int = 2
+    """Number of neighbors to retrieve during pretraining."""
+
+    retro_num_retrieved_chunks: int = 2
+    """Number of chunks to retrieve from the retrieval database."""
+
+    retro_retrieved_length: int = None
+    """Cached value of retro_num_retrieved_chunks * retro_chunk_length (i.e., the total number of
+       retrieved tokens; neighbor + continuation).
+    """
+
+    retro_split_preprocessing: str = None
+    """Data split used during data preprocessing."""
+
+    retro_verify_neighbor_count: bool = True
+    """Verify that len(GPT dataset) == len(saved neighbors)."""
+
+    def __post_init__(self) -> None:
+        """Validate Retro config."""
+
+        super().__post_init__()
+
+        # Validate Transformer Engine version.
+        te_version = packaging.version.Version(version("transformer-engine"))
+        if te_version >= packaging.version.Version("1.3"):
+            try:
+                assert os.getenv("NVTE_FLASH_ATTN") == "0"
+                assert os.getenv("NVTE_FUSED_ATTN") == "0"
+            except Exception as e:
+                raise Exception(
+                    "When using Transformer Engine >= 1.3, environment vars NVTE_FLASH_ATTN and NVTE_FUSED_ATTN most both be defined and set to '0'. Currently, NVTE_FLASH_ATTN == %s, NVTE_FUSED_ATTN == %s."
+                    % (
+                        os.getenv("NVTE_FLASH_ATTN", "[unset]"),
+                        os.getenv("NVTE_FUSED_ATTN", "[unset]"),
+                    )
+                )
+
+        # Preprocessing split should be defined.
+        assert self.retro_split_preprocessing is not None
+
+        # Pre-compute retrieved length.
+        self.retro_retrieved_length = self.retro_num_retrieved_chunks * self.retro_chunk_length